Get into My Car(-T)

During a workshop event at Advanced Therapies Week in Miami in January 2024, we asked, what if we could produce CAR-T (or any other) cell therapies using highly matured methods from other industries, such as automotive manufacturing? The session “Cell Therapy Manufacturing: A Look Under the Hood – Learning from Outside the Industry,” sparked interesting conversations with attendees.

Existing methods for producing CAR-T’s are hampered by a lack of standardized sterile liquid handling steps, leading to inefficiencies and inconsistencies. There are a variety of cell culture vessels that are challenging to scale up, including culture bags, flasks, cell stacks, and big tank bioreactors – and these may be open or closed, requiring either static or dynamic incubation, with multiple media exchanges or none. Though an artisanal approach has produced groundbreaking clinical results in the small population of patients able to receive them, the industry will struggle to manufacture enough therapies to meet the soaring demand.

Automotive insights
 

In all industries, new technologies go through multiple cycles of continuous improvement before they become stable, predictable, and more affordable. Most of these new technologies are developed from a belief that everything must be created from scratch using a trial-and-error approach, which is very expensive.

At first glance, the automobile industry’s best practices may seem very far from the healthcare industry, but it is actually the exemplar of success using continuous improvement methodologies. The key is an unyielding commitment to key strategies, such as lean manufacturing – a systematic approach developed by Toyota in the mid-20th century to maximize productivity, which focuses on streamlining workflows and reducing waste. By transplanting some of these concepts into the realm of cell therapies, we can reimagine a process that maximizes efficiency and ensures consistent product quality.

One possibility is assembly line manufacturing, a pillar of continuous improvement that the automotive industry pioneered over 100 years ago. For much of the cell therapy industry today, manufacturing is limited by slots – one therapy is made at a time, and a new one can’t enter the system for manufacturing until the first one exits. In part, this concept is driven by the often lengthy cell expansion phase. But if we break the cell therapy manufacturing process into parts or intervals of time, however, it could enable customized scheduling to maximize critical asset utilization.

During the workshop, we demonstrated that the cell therapy production process can be segmented into discrete units for each day by providing slots of times to every process step, using the 75th percentile of all historical time durations to execute that process step. Those steps result in an assembly line process that can be fixed to a schedule in a predictable manner. With such a level of specificity, it becomes possible to increase capacity by three- to fivefold.

Assembly line production enables the integration of many other lean tools for continuous improvement and structural cost reduction. These include task-driven organization to accelerate operator training and qualification, “takt-time” to predict throughput, and “level loading” to maximize asset utilization for operators, incubators, and biosafety cabinets. 

Putting the CAR-T before the car
 

Lean manufacturing works hand-in-hand with automation to reduce costs. Based on the way it has been successfully implemented in other industries, we believe the cell therapy space is largely going about it backwards – developing expensive, complex equipment to automate before processes are optimized. Robots aren’t known for their flexibility. When automation is used too soon, it hinders the ability to pivot if things don’t work the way they are supposed to. As a result, developers transitioning to such equipment may wind up locking wasteful processes into a new automated standard.

Starting from lean methodology gives developers a chance to first define then optimize a standard, enabling the use of simpler and more cost-effective tools before exploring automation. Only once a process is well-optimized should automation be introduced – either totally (end-to-end) or partially as standardization moves along each step.

In lean methodology, there is a term for a big machine meant to automate a process but implemented too soon in the optimization journey: a monument. Eventually, obsolete and unused, it becomes very expensive real estate in a lab or cleanroom – and enormously costly to dispose of.

To repeat: simplify, define, and optimize standards first, and then consider automation. Don’t get stuck with a monument!

Most people in our industry are primarily concerned about a lack of capacity. Surprisingly, when we optimize with assembly lines, consolidate technology, and use more powerful metrics, it becomes clear there is actually a huge manufacturing underutilization issue that should be our focus instead. Although it is hard to pin down, a conservative extrapolation based on process mapping of CAR-T, TCR, TILs or NK cells tells us that manufacturing facilities are used less than 40 percent of the time.

The simplified standard
 

There are more than 100 years of learnings that are applicable to cell therapy manufacturing from more closely related industries than automotive, including food, medical devices, and traditional pharma. Such cross-industry innovation can be observed in the simplified tools that are gaining popularity in cell therapy manufacturing. Platforms such as flexible cell manufacturing devices, for example, enable the process to flow seamlessly across process steps and machines.

Because these flexible manufacturing devices lack the complexity of large-scale automated platforms, they can be more easily integrated into end-to-end manufacturing for different cell therapy platforms. They should also enable both autologous and allogeneic modalities to use the same core bioreactor tech. As these cell culture devices become standard, they enable other medical devices and service providers to adapt their offerings to support even higher production efficiency – for example, wider incubators or fill/finish systems for multiple bags or faster CMC document generation based on predictable and stable technology.

The use of these devices also simplifies metrics to compare costs across the industry. Much like the Big Mac Index, which uses the cost of assembly line-produced hamburgers to assess cost of living in different countries, it is possible to make economic decisions by breaking down and assessing cell therapy in discrete steps. We use the total cost of utilizing an enabling device – including labor, materials, and overheads – versus throughput, measured in billions of cells per device. This type of metric allows the cell therapy industry to compare technologies, internally or across multiple CDMO sites, and make educated decisions. It can help avoid spending millions of dollars in technologies, partnerships, or pathways that won’t yield the expected results. Once the cell therapy process is stabilized, the same metric can be applied to help calculate the cost per patient dose. This approach can also help manufacturers understand the intrinsic process variation coming from the cell biology associated with these individual technologies, as well as increase utilization knowhow. These factors play a critical role in reducing costs while increasing yields, enabling developers to do more with less, as production costs with a single, simple device will always be stable relative to the company’s own technology.

By borrowing from the playbook of other successful industries, we can start a new era in cell therapy manufacturing – one marked by efficiency, scalability, and a newfound capacity to meet the global need for these transformative therapies. By integrating principles such as lean manufacturing and by adopting technological advancements, we can transform cell therapy production from an artisanal process to a streamlined, high-throughput system.